Behavioral thermoregulation in a gregarious lemur Eulemur collaris Effects of climatic and dietary-related factors.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 144:355–364 (2011) Behavioral Thermoregulation in a Gregarious Lemur, Eulemur collaris: Effects of Climatic and Dietary-Related Factors Giuseppe Donati,1,2* Eva Ricci,2 Nicoletta Baldi,2 Valentina Morelli,2 and Silvana M. Borgognini-Tarli2 1 Nocturnal Primate Research Group, Department of Anthropology and Geography, Oxford Brookes University, Oxford, OX3 0BP, UK 2 Department of Biology, Unit of Anthropology, University of Pisa, Pisa, Italy KEY WORDS Madagascar Eulemur collaris; thermal ecology; nutritional ecology; canonical correlation; ABSTRACT Primates deal with ﬂuctuations of the thermal environment by both physiological and behavioral mechanisms of thermoregulation. In this article we focus on nonhibernating lemurs, which are hypometabolic and have to cope with a seasonal environment. Behavioral thermoregulation has received little attention compared with speciﬁc physiological adaptations to seasonality, i.e., hibernation and torpor, which characterize a number of lemurs. We investigated the role of seasonality and dietary-related factors in determining frequencies of resting, social and postural thermoregulation, and microhabitat selection in collared lemurs, Eulemur collaris. We observed two groups of collared lemurs over a 14-month period in the littoral forest of Sainte Luce, Southern Madagascar. Frequencies of total resting and time spent in huddling, prone, and curled postures were collected via 5-min instantaneous sampling. Microhabi- Endothermal animals use a series of behavioral strategies to supplement autonomic physiological mechanisms of thermoregulation, including activity and postural adjustments, social thermoregulation, and microhabitat selection (Reﬁnetti, 1998; Seebacher and Franklin, 2005). The ﬁrst two strategies include saving energy by reducing activity and controlling heat-radiation by modifying the body surface/volume ratio, while the latter consists of the active search for a microhabitat whose thermal conditions approach the thermoneutral zone (Stevenson, 1985), i.e., the range of environmental temperatures over which the heat produced by an endotherm remains fairly constant. In addition to chemical thermogenesis and the control of peripheral circulation, behavioral thermoregulation represents an important system to reduce energy expenditure for thermoregulation, and, in many species, to minimize body temperature variations in case of small thermal ﬂuctuations. In mammals supplementary physiological mechanisms (e.g., shiver or perspiration) are activated when behavioral adaptations are insufﬁcient or inadequate (Satinoff, 1978). Among mammals, both nonplacental semi-poikilotherms (McNab, 1978; Rübsamen et al., 1983; Körtner and Geiser, 2000; McCarron et al., 2001; Brice et al., 2002; Grigg et al., 2003; Bethge et al., 2004) and placental endotherms (Bradley and Hudson, 1974; Hasler and Sorenson, 1974) use the full set of behavioral thermoregulatory strategies, although in different combinations and in different proportions. Primates have been shown to resort to microhabitat selection, adjustments in posC 2010 V WILEY-LISS, INC. tat selection was evaluated as the proportion of time spent in the upper canopy as compared with other layers. Climatic variables were recorded by automatic data loggers, while dietary variables were derived from phenological data and nutritional analyses of the ingested food items. We weighted the combined effects of climatic and dietary variables on the different types of behavioral thermoregulation by means of canonical correlation analysis. The model with the strongest canonical correlation included a ﬁrst root representing mainly feeding time, day length, and ambient temperature and a second root representing diet quality and height of feeding trees. The output indicated that collared lemurs adapt to thermal and dietary-related metabolic stress by adjusting resting time, social, and postural thermoregulation. Am J Phys Anthropol 144:355–364, 2011. V 2010 C Wiley-Liss, Inc. ture and activity, and huddling as responses to environmental temperature (Microcebus murinus, Aujard et al., 1998; Varecia variegata, Morland, 1993; Leontopithecus rosalia, Thompson et al., 1994; Alouatta caraya, BiccaMarquez and Calegaro-Marquez, 1998; Colobus polykomos, Dasilva, 1993; Papio cynocephalus, Stelzner and Hausfater, 1986; Pochron, 2000; Hill and Dunbar, 2002; Hill et al., 2004; Hill, 2006; Macaca fuscata, Schino and Troisi, 1990; Hanya et al., 2007; Cebus capucinus, Campos and Fedigan, 2009; Pan troglodytes, Takemoto, 2004; Kosheleff and Anderson, 2009). These studies demonstrate that behavioral thermoregulation represents an important dimension in primate Grant sponsors: Department of Biology, University of Pisa; Department of Anthropology and Geography, Oxford Brookes University; Department of Animal Ecology and Conservation, University of Hamburg. *Correspondence to: Giuseppe Donati, Nocturnal Primate Research Group, Department of Anthropology and Geography, Oxford Brookes University, Oxford, OX3 0BP, United Kingdom. E-mail: firstname.lastname@example.org Received 1 December 2009; accepted 27 August 2010 DOI 10.1002/ajpa.21415 Published online 26 October 2010 in Wiley Online Library (wileyonlinelibrary.com). 356 G. DONATI ET AL. time-budgets. Although relevant literature is available on social thermoregulation in small mammals (Kauffman et al., 2003; Kotze et al., 2008), we do not know much about the relative importance of different forms of behavioral thermoregulation as a reaction to the thermal environment in primates. A further problem is to identify the role of climatic factors affecting the perception of external temperature, among them relative humidity, wind velocity, and solar radiation (Hill et al., 2004). A high relative humidity, for instance, increases perceived temperature, because of a decrease in evaporation rate, while perceived temperature decreases in the presence of wind because of the higher efﬁciency of heat loss by convection (Mount, 1979). Also, behavioral thermoregulation of wild mammals has been mainly analyzed in relation to climatic variables, while other ecological aspects, such as dietary-related factors, rarely have been considered. In fact, nonclimatic variables, such as food availability and quality, as well as social factors, may have thermoregulatory implications (Dhal and Smith, 1985; Schino and Troisi, 1990; Dasilva, 1993). Malagasy lemurs show an extensive set of behavioral, postural, and social activities which seem to function as thermoregulatory mechanisms (Richard and Dewar, 1991; Morland, 1993; Ostner, 2002). This has been interpreted as a response to the pronounced seasonality and climatic unpredictability that characterize the Malagasy island environment (Jolly, 1984; Wright, 1999; Dewar and Richard, 2007). The lemurs studied so far appear to be hypometabolic (Genoud, 2002; Simmen et al., 2010) with limited sweating mechanisms (Aujard et al., 1998), and it has been suggested that they should often resort to behavioral thermoregulation (Morland, 1993). In particular, E. fulvus, a sister species of E. collaris, object of this research, has been shown to have a basal metabolic rate which is among the lowest in prosimians, ranging from 28 to 56% of the expected value of the Kleiber equation, coupled with a relatively constant body temperature (Daniels, 1984; van Schaik and Kappeler, 1996). Therefore, it is reasonable to hypothesize that Eulemur spp. will have a wide range of mechanisms to ensure temperature homeostasis, including behavioral strategies. The aim of this research was to investigate the role of climatic and dietary-related factors on year-round behavioral thermoregulation in two groups of collared lemurs living in a littoral forest fragment of South-eastern Madagascar. We ﬁrst examined whether seasons play a role in affecting activity adjustments, social and postural thermoregulation, and microhabitat selection separately. For this, we compared between groups and among seasons the percentage of time spent resting (activity adjustments), huddling (social thermoregulation), resting in curled and prone positions (postural thermoregulation), and time spent in the upper canopy (microhabitat selection). If seasonality is inﬂuential, we predicted that lemurs would rest more and use huddling and curled postures more frequently during cold seasons and would spend more time in prone postures during warm periods. We also expected lemurs to spend more time in the lower layers of the forest during warm periods to get cooler and use the upper levels more to warm up during cold seasons. We further examined the role of seasonality by modeling the simultaneous role of some climatic and dietary-related factors in determining frequencies of the various types of behavioral thermoregulation in a multivariate analysis. Given the overall resource seasonality of Madagascar, we predicted that lemur thermoregulaAmerican Journal of Physical Anthropology Fig. 1. Average monthly temperature and rainfall at Sainte Luce during the study period. Seasonal segments are indicated. Tra-dry: transitional dry season. tory behaviors would be shaped not only by climatic, but also by dietary-related factors: when food is scarce or has lower energy content lemurs should use energy saving behaviors more often. MATERIALS AND METHODS Study site and study species Observations were conducted between December 1999 and January 2001 at the Sainte Luce Conservation Area by GD, NB, and VM in a 377 ha fragment of the littoral forest (24845’s 478110 E), 30 km north of Fort Dauphin (South-eastern Madagascar). This forest, occurring within 2–3 km from the coast at an altitude of 0–20 m a s l, is characterized by a tropical wet climate, with temperatures ranging from 15 to 308C and total annual rainfall of 2,480 mm. These littoral forests have a relatively open or noncontinuous canopy, which is 6–12 m in height with emergents up to 20 m (Bollen and Donati, 2005). Though the area is characterized by an overall aseasonal climate, during the study period four climatic segments were identiﬁed: hot-wet (December–February), hotdry (March–May), cool-wet (June–August), and transitional-dry (September–November; see Fig. 1; Donati et al., 2007). A month was considered wet when total rainfall was above 100 mm (Morellato et al., 2000), cool when average 24-h temperature was below 218C (see Fig. 1). Collared lemurs are cathemeral medium-sized primates (head-body length: 39–40 cm, tail length: 50–55 cm, weight 2.3–2.5 kg). They are mostly frugivorous and live in multimale-multifemale groups ranging from 3 to 17 individuals (Donati et al., 2007). Behavioral data Two groups of collared lemurs (Group A size: 8–13 individuals; Group B size: 4–6 individuals) were followed 3 days each month from 6:00 am to 6:00 pm. Group A was followed during 13 months; Group B during 10 months, totaling 948 observation hours. Individual identiﬁcation was realized using nylon collars and colored pendants, and one individual per group was radio-collared. Animal activity was recorded by the instantaneous focal method (Altmann, 1974) at 5-min intervals. Focal animals were chosen from all adult individuals in both study groups, and were rotated every 3 h, so that at the end of 3 observation days (12 observation h day21) all adult group members had been evenly sampled. Truncated observations (very rare) were discarded when covering less than 70% of the respective focal session. Active behaviors (feeding, foraging, mov- BEHAVIORAL THERMOREGULATION IN COLLARED LEMURS ing, and social behaviors), nonactive behaviors (resting alone or huddling with one or more others), food type consumed (mature and unripe fruits; mature and young leaves; nectar; ﬂowers; animal matter; other), and level above ground (measured at 2-m intervals) were registered during observations. The daily proportion of huddling with another animal weighted by the total number of instantaneous records was used to quantify social thermoregulation. Resting postures were categorized as follow: 1) hunch: the animal curled like a ball with the tail wrapped around; 2) curl: as position 1 but with the tail hanging down; 3) sit: stationary posture when the animal is upright; 4) ventral prone: the animal lays on abdomen with limbs and tail hanging down; 5) dorsal prone: the animal lays on back with limbs and tail hanging down. Postural thermoregulation was quantiﬁed using two variables: the daily proportion of instantaneous observations in postures 1 and 2, ‘‘curled’’ postures, and the daily proportion in postures 4 and 5, ‘‘prone’’ postures. We excluded posture 3 from the analysis because its signiﬁcance for thermoregulatory function was affected by the fact that this posture is used during activities such as feeding, grooming, inspecting, etc. When taken alone posture 3 was often transitorily maintained for very short periods, so it cannot be considered a purely ‘‘resting posture.’’ Microhabitat selection was quantiﬁed as the daily proportion of records in the canopy, i.e., above 6 m. This is the lower limit of the forest canopy and the animals were on average more exposed to meteorological factors when resting/moving above this height. In a previous study, we recorded appreciable differences in terms of daily temperature and relative humidity between the understory of the littoral forest and the canopy, i.e., larger ﬂuctuations characterize the upper levels (up to 18C for temperatures and up to 10% for relative humidity; Donati and Borgognini, 2006). Climatic data 357 Donati, 2005). The overall availability was obtained by weighting the presence of a speciﬁc phenophase for each species by its mean DBH. Diameter at breast height has been demonstrated to be a reliable proxy of the quantity of fruits produced by a tree (Ganzhorn, 2003). All feeding trees used by the animals for more than 5 s were marked and identiﬁed via the help of a local expert. On these occasions, tree height was measured with a clinometer or estimated to the nearest meter by eye. Nutritional analyses were performed at the Department of Animal Ecology and Conservation of Hamburg University. A total of 112 food samples representing all the species eaten by collared lemurs during the study period were analyzed (in some cases one among the transect 78 species gave more than one food item, e.g., leaves, fruits, ﬂowers, and some food species not present in the phenological transects were also analyzed). Fresh and dry weight was determined for each sample before the analyses. Lipid content was determined by petroleum/ether extraction. Protein content was estimated by the Biorad procedure. Soluble carbohydrates were assessed after extraction with 50% methanol as the equivalent of galactose after acid hydrolization (Ortmann et al., 2006). Neutral detergent ﬁber content (NDF) was determined according to van Soest (1994), modiﬁed according to the instructions for use in an ‘‘Ankom ﬁber analyzer.’’ To evaluate diet quality, we calculated the metabolizable energy in the monthly diet. We calculated the weighted proportion of dry matter per month for each nutritional compound, with the proportion of feeding records for each food item as the weighted coefﬁcient (Kurland and Gaulin, 1987). Energy content from food was obtained by standard conversion factors such as 4 kcal g21 for carbohydrate, 4 kcal g21 for crude protein and 9 kcal g21 for lipid. We used a ﬁber conversion factor of 3 kcal g21 rather than 4 kcal g21 usually used for carbohydrates, since the anaerobic microbes take 1 kcal g21 of ﬁbers for their own growth during fermentation processes (Conklin-Brittain et al., 2006). The metabolizable energy was then obtained via the following equation: Temperature and relative humidity were registered at 2-h intervals by data loggers, Hobo H8 pro, operated by custom software (BoxCar 3.51 for Windows, Onset Computer Corporation). Two data loggers were positioned at 1 and 10 m above ground level at the boundary between the home ranges of the two study groups. Rainfall was measured every day at 06.00 h using a rain gauge placed near the camp. Daily data on wind velocity were registered by a station located at the Fort Dauphin airport and kindly provided by the Direction Générale de la Météorologie of Antananarivo. Solar radiation was not considered in this study because of its believed limited importance for forest-dwelling species (Hill et al., 2004). where ME is the metabolizable energy per gram (in kcal g21) of diet; L is the proportion of lipids; SP the proportion of soluble proteins; SC the proportion of soluble carbohydrates and [NDF 3 0.415] the fraction of NDF which are digested by brown lemurs (Campbell et al., 2004). Dietary-related data Data analyses To estimate variation in food availability, phenological data were recorded for the plant species (n 5 78) whose fruits were consumed by collared lemurs during the study period. Diameter at breast height (DBH) was measured for all the species used by the lemurs as feeding trees in two botanical transects covering 2,320 m 3 10 m. The ﬁrst ﬁve adult (DBH 10 cm) individuals for each tree species were selected in each transect to collect phenological data. Trees were checked for the presence/absence of fruits twice a month (Bollen and The records of thermoregulatory activity were weighted by the total number of instantaneous records. Daily proportions were calculated for each group and these data were log transformed to allow the use of multivariate parametric tests. As a ﬁrst step, we evaluated the inﬂuence of seasons in determining daily proportions of each thermoregulatory behavior via a two-way repeated measure ANOVA. Within-subjects factors were the two groups and the four seasonal segments (hot-wet, hot-dry, cool-wet, and transitional-dry). ME ¼ ð9 3 LÞ ð4 3 SPÞ þ ð4 3 SCÞ þ ð3 3 ½NDF 3 0:415Þ American Journal of Physical Anthropology 358 G. DONATI ET AL. Second, we estimated the association between climatic and dietary-related factors and the proportions of total resting, huddling, postural thermoregulation, and microhabitat selection via Canonical Correlation Analysis (CANCOR). CANCOR is suited for the mathematical description of situations in which there are two sets of variables obtained under the same conditions and the objective is to study the relationship between these two sets. This technique forms pairs of weighted linear combination of variables (latent roots), one in each set, such that the correlation between the pair of latent roots is maximized (McGarigal et al., 2000). The ﬁrst set of variables, i.e., the predictor ones, included a group of climatic factors: daily average ambient temperature and daily average relative humidity (both averaged over six data points, as we used only observations from the 12 daylight hours), max daily wind speed, daily rainfall, day length; and a group of dietary-related factors: monthly metabolizable energy in the diet, daily proportion of feeding time, monthly fruit availability, monthly average height of feeding trees (calculated averaging the heights of all the trees used to feed during a given month). The second set of variables, i.e., the dependent ones, included: daily resting time (activity adjustments), daily proportion of time in huddling (social thermoregulation), daily proportion of time in prone postures and daily proportion of time in curled postures (postural thermoregulation), daily proportion of time spent in the upper canopy (microhabitat selection). Indices of perceived temperature were not included in the present analysis because of their high correlation (Pearson: r [ 0.8) with the original climatic variables and the consequent violation of the CANCOR model. We included feeding time in the independent data set and resting time in the dependent data set because of the higher biological priority of the former on the latter (sensu Hill et al., 2003). All the variables used in the multivariate model were numerical and entered the analysis after log or arcsin transformation to improve their linearity. The CANCOR model accepts variables measured with different scales, i.e., both interval and ordinal variables. RESULTS Effect of seasons Activity adjustments. During the study period Groups A and B spent 53.4% 6 11.3% and 58.7% 6 9.9% of their time resting, respectively (overall 56.1% 6 11.2% of the total instantaneous records, N 5 11,376; Fig. 2a). Over the four seasons, Group B reached a maximum of 66.3% 6 5.1% of its time inactive during the cool-wet season, while the minimum was recorded in Group A during the hot-wet season (41.5% 6 8.5%). A signiﬁcant group effect (F1,60 5 34.92, P 5 0.027) and an effect of seasons (F3,60 5 6.89, P 5 0.023) were observed. Social thermoregulation. Overall, the two groups of collared lemurs spent 31.3% 6 17.6% of their time in huddling, with Group A devoting 29.4% 6 17.7% and Group B 33.7% 6 17.3% of their time to this behavior (Fig. 2b). Over the four seasons, the maximum time spent in huddling was observed in Group A during the transitional-dry season (44.6% 6 13.7%), while the minimum was recorded in the same group during the hot-wet season (18.3% 6 13.9%). Only a major effect of seasons was revealed (F3,65 5 4.49, P 5 0.025). American Journal of Physical Anthropology Postural thermoregulation. During the study period the two groups spent 37.6% 6 16.7% and 6.4% 6 9.9% of their time in curled and prone postures, respectively (Fig. 2c,d). Curled postures peaked during the transitional-dry season in Group A (49.5% 6 14.4%) and reached their minimum during the hot-dry season in Group B (20.6% 6 16.9%). Conversely, prone postures were observed more frequently during the hot-dry season in Group B (14.9% 6 14.5%), while they were almost absent during the cool-wet season in Group A (0.2% 6 0.4%). A strong tendency to a seasonal effect (F3,65 5 3.41, P 5 0.053) and an interaction effect between groups and seasons (F3,65 5 4.25, P 5 0.029) were found for time spent in curled postures. Seasonal and group effects were nonsigniﬁcant for time spent resting in prone postures. Microhabitat selection. The two groups spent 38.5% 6 27.1% of their time in the upper canopy (Fig. 2e). Group A spent a greater proportion of time in the highest forest layers, 45.8% 6 24.3%, than Group B, 28.9% 6 28.0%. The animals were observed more frequently in the upper canopy during the hot-dry season (Group B: 63.0% 6 19.8%), while this was infrequent during the hot-wet season (Group B: 16.8% 6 26.3%). Time spent in the higher part of the canopy and in the emergents did not change signiﬁcantly between the two groups and over the four seasons. Effect of climatic and dietary-related factors Descriptive statistics showing the seasonal variation of the nine variables represented by climatic and dietaryrelated seasonal factors (independent variables) and the ﬁve variables represented by the four forms of behavioral thermoregulation (dependent variables) are illustrated in Table 1. The ﬁrst two pairs of latent roots derived from the canonical correlation between the set of independent variables and the set of dependent variables were highly signiﬁcant (First root: v2 5 170.20, df 5 45, P \ 0.001, R2 5 89.30%; Second root: v2 5 66.28, df 5 32, P \ 0.001, R2 5 56.10%). The redundancy analysis showed that the ﬁrst two pairs of latent roots together were able to explain 56.61 and 34.21% of variance in the dependent and independent sets of variables, respectively. Successive removal of latent roots did not give further signiﬁcant inter-correlations. To understand the relationship between the original variables and the composite variables created by the CANCOR model, the matrix of correlations between the ﬁrst two pairs of roots and the original variables is shown in Table 2. The strength of the association between each original variable and the roots is given by the value of the correlation. Summarizing, the ﬁrst root shows that as feeding time, ambient temperature, day length, fruit availability, and height of feeding trees decrease, resting time and proportion of resting in huddling and in curled postures increase, while time spent in prone posture decreases. The second root shows that as quality of the diet and height of feeding trees decrease, the proportion of resting in huddling and in curled postures increases, while time spent in the upper canopy decreases. To visualize in two dimensions the relationships among the above variables, we plotted the score of each observation day in the new space created by the two latent roots in each set of variables (see Fig. 3). BEHAVIORAL THERMOREGULATION IN COLLARED LEMURS 359 Fig. 2. Mean percentage of time spent resting (a) and mean percentage of time spent in huddling (b), in curled postures (c), in prone postures (d), and in the upper canopy (e) by the two study groups over the four seasonal segments. Tra-dry: transitional-dry season. Error bars represent standard deviations. DISCUSSION The role of seasons The tight relationship between time allocated to resting, resting postures and seasonal changes seems to be a paramount thermoregulatory strategy and has been repeatedly observed in studies on lemurs (Morland, 1993; Pereira et al., 1999), on other primates (Stelzner and Hausfater, 1986; Schino and Troisi, 1990; Dasilva, 1993; Thompson et al., 1994; Bicca-Marquez and Calegaro-Marquez, 1998; Pochron, 2000; Hill et al., 2004; Hill, 2006; Hanya et al., 2007; Campos and Fedigan, 2009; Kosheleff and Anderson, 2009), and on mammals in general (Bradley and Hudson, 1974; Hasler and Sorenson, 1974; Körtner and Geiser, 2000). Our results show that seasonal ﬂuctuations had a major role in shaping the four types of behavioral thermoregulation used by collared lemurs. As predicted, resting time and its proportion spent in postures which reduce the surface/volume ratio, i.e., huddling and curling, increased during the colder seasons indicating that these strategies were used to minimize the dissipation of body heat. By contrast, the proportion of time spent in prone, extended postures was very low during the colder part of the year and increased during the warm season, though these differences were not statistically signiﬁcant. This is also a predicted result, since, following the same principle of surface/volume ratio body; elongation maximizes dissipation of heat when the thermal environment is particularly warm. Thus, together with activity adjustments and postural thermoregulation, social thermoregulation appears to have a relevant role in dealing with American Journal of Physical Anthropology 360 G. DONATI ET AL. TABLE 1. Means, standard deviations, and sample size of the 14 climatic, dietary-related, and behavioral variables used in the canonical correlation model Independent variables Seasons AT Climatic factors DL RH RF Hot-wet SD n Hot-dry SD n Cool-wet SD n Tra-dry SD n 25.3 1.0 20 24.9 1.3 14 19.7 1.1 18 22.7 1.7 13 13.5 0.2 20 11.4 0.5 14 10.9 0.3 18 12.6 0.7 13 93.3 3.9 17 89.9 7.0 10 96.1 2.7 15 89.5 6.1 13 3.1 4.8 17 2.6 6.4 10 13.6 31.1 15 0.6 0.7 13 WS FT 12.1 3.2 17 14.2 4.6 10 15.0 4.8 15 18.0 5.5 13 27.4 5.2 20 21.4 6.3 14 19.2 2.4 18 18.7 1.3 13 Dependent variables Dietary-related factors ME FA HF 1.6 0.2 6 1.9 0.2 4 1.8 0.3 6 1.5 0.2 5 355.1 91.0 5 201.7 72.6 3 151.8 45.5 3 270.9 23.6 3 9.3 0.9 5 9.9 0.4 3 8.3 1.0 6 9.9 0.5 5 RT Behavioral thermoregulation HU CP PP UC 51.2 6.4 20 59.8 8.0 20 69.1 7.0 18 66.3 7.9 13 23.8 17.9 20 21.4 14.5 20 40.2 12.0 18 41.1 16.4 13 27.9 14.5 20 32.1 17.8 20 45.5 9.8 18 47.7 16.5 13 10.2 11.8 20 9.6 11.4 20 3.5 7.5 18 1.2 3.2 13 34.4 28.8 20 56.4 19.3 20 35.1 29.5 18 30.5 21.8 13 AT: ambient temperature (8C); DL: day length; RH: relative humidity (%); RF: rainfall (mm); WS: wind speed (meters per second); FT: feeding time (% of total records); ME: metabolizable energy in the diet (kcal); FA: fruit availability; HF: height of feeding trees (meters); RT: resting time (% of total records); HU: resting in huddling; CP: resting in curled postures; PP: resting in prone postures; UC: time in the upper canopy; Tra-dry: transitional dry season. TABLE 2. Pearson correlations between the ﬁrst and the second latent roots derived from the canonical correlation and the original variables in the two datasets Independent set 1st Root 2nd Root Dependent set 1st Root 2nd Root Ambient temperature Day length Energy in the diet Feeding time Fruit availability Height of feeding trees Relative humidity Rainfall Wind speed 0.703 0.669 20.006 0.886 0.474 0.401 0.069 20.035 20.129 0.132 0.298 20.607 0.088 0.193 20.465 20.328 0.113 0.251 Resting in huddling Resting in curled postures Time in upper canopy Resting in prone postures Resting time 20.575 20.720 0.166 0.395 20.993 0.391 0.431 20.682 20.086 0.029 Signiﬁcant correlations after Bonferroni adjustment are indicated in bold. seasonal ﬂuctuations. Only a few studies have shown the association between huddling frequencies and seasons in gregarious lemur species (Jolly, 1966; Tattersall, 1982; Pereira et al., 1999; Ostner, 2002). However, social thermoregulation seems to play an important role even for solitary ranging, nocturnal lemurs, which form day-time sleeping groups (Fietz and Dausmann, 2006). A large amount of work has provided evidence that social thermoregulation can be a source of substantial energy savings, since lower metabolic rates due to huddling have been demonstrated in many species of small mammals (Kauffman et al., 2003; Kotze et al., 2008). It has been hypothesized, for example, that one of the reasons why African four-striped grass mice may sometimes occur in groups is that energetic beneﬁts can be gained through huddling in habitats in which food and water are scarce (Scantlebury et al., 2006). The importance of social thermoregulation in lemurs has also suggested the hypothesis that the unusual sex ratio found in brown lemur groups, equal or male-biased (Kappeler, 2000), might be advantageous for possible thermoregulatory beneﬁts given by extra males (Overdorff, 1998; Ostner, 2002). Our data collection was not designed to test this hypothesis, but in the larger study group, Group A, males outnumbered females by 2:1. If animals huddle in proportion to their numbers, we would expect a greater huddling frequency in Group A. However, we did not notice a tendency of the more numerous group, where several subordinate males were present, to participate more often in huddling than the smaller group. American Journal of Physical Anthropology Prone or upright body postures in sunny spots with arms extended laterally are also used by primates to gain heat (Jolly, 1966; Takemoto, 2004). Sunning or basking in early morning is a common behavior in ringtailed lemurs, sifakas, and ruffed lemurs (Jolly, 1966; Tattersall, 1982; Morland, 1993; Pereira et al., 1999). Among anthropoids, wild Japanese macaques spend a signiﬁcant amount of time resting on the trees during the winter to warm up (Hanya et al., 2007). However, our data indicate that collared lemurs did not use basking or sunning as a thermoregulatory strategy in the littoral forest, and lying, prone postures were almost absent during the winter. In a primary forest the only opportunity for sunbathing is on the emergent part of the canopy, where predation risk by diurnal raptors is expected to be high. The absence of sunning behaviors in collared lemurs is in line with other ﬁeld studies on Eulemur species (Sussman, 1974; Curtis et al., 1999; Pereira et al., 1999). Apparently, these taxa rely on other strategies to warm up early in the morning. One of these tactics could be the frequent use of social thermoregulation and curled postures, as this study seems to suggest. Interestingly, the more egalitarian society and the high degree of tolerance among brown lemurs as compared to other lemurs (Kappeler, 1993; Pereira et al., 1999) could favor social thermoregulation. Cathemerality has also been suggested as a possible strategy to warm up by being active during the coldest hours of the day, i.e. before sunrise, in Eulemur species (Curtis et al., 1999). BEHAVIORAL THERMOREGULATION IN COLLARED LEMURS 361 Fig. 3. Biplots of the scores of the observation days on the axes of the ﬁrst latent root (a) and on the axes of the second latent root (b) derived from the canonical correlation analysis. Abbreviations represent the original variables most associated with the new canonical axes. The font size is in proportion to the strength of the correlation between the original variables and the canonical roots. The arrows indicate the direction of increase of each original variable. AT: ambient temperature; DL: day length; FT: feeding time; ME: metabolizable energy in the diet; FA: fruit availability; HF: height of feeding trees; RT: resting time; HU: resting in huddling; CP: resting in curled postures; PP: resting in prone postures; UC: time in the upper canopy. Open squares: hot-dry season; open diamonds: hot-wet season; open triangles: cool-wet season; open circles: transitional-dry season. Physiological data are necessary to test the latter hypothesis, but it seems that cathemeral activity in these lemur species is triggered by multiple factors (Curtis and Rasmussen, 2006; Donati et al., 2009). As to differences between groups, our results also show that Group A had signiﬁcantly lower levels of resting than Group B. Group A occupied an interior, central area of the fragment where the forest was relatively undisturbed. Conversely, the home range of Group B was located in the southern portion of the fragment which was characterized by frequent use by local villagers due to the proximity to the nearby settlement and the national road. The structural proﬁle of these two areas has been measured by the line-intersect technique and signiﬁcant differences in canopy cover were recorded (Rakotondranary, 2004; Ganzhorn et al., 2007). If these structural differences have resulted in micro-climatic differences, the higher level of resting of Group B might be a general strategy to save energy in a more variable thermal environment. Forest degradation and edge effects have been repeatedly shown to modify densities, ecology, and behavior of primates (Lovejoy et al., 1986; Ganzhorn, 1995; Estrada and Coates-Estrada, 1996; Onderdonk and Chapman, 2000; Lehman et al., 2006; Rode et al., 2006). In degraded and logged areas animal time-budget is altered and an increase of resting is often observed (Johns, 1986; Cowlishaw and Dunbar, 2000). However, climatic data from both areas and behavioral records on a larger number of animals are a prerequisite to verify the effects of such micro-climatic changes in the Sainte Luce littoral forest. The role of climatic and dietary-related factors A relevant problem in examining further the role of seasonal ﬂuctuations in shaping behavioral thermoregulation is the difﬁculty in accounting for the variety of factors which inﬂuence perceived temperatures and for the effect of nonclimatic factors. As for climatic factors, ambient temperature, humidity, solar radiation, and wind speed have all been shown to inﬂuence behavior in baboons (Stelzner and Hausfater, 1986; Pochron, 2000; Hill and Dunbar, 2002; Hill et al., 2004; Hill, 2006), in howler monkeys (Bicca-Marquez and Calegaro-Marquez, 1998), in capuchins (Campos and Fedigan, 2009), and in chimpanzees (Kosheleff and Anderson, 2009). An index of perceived environmental temperature that accounts for the variation of climatic factors has also been proved to be effective in predicting baboon’s resting and feeding behavior (Hill et al., 2004). However, we did not use an index of perceived temperature in our analysis because of the unknown bias in applying an index developed for humans to a species with very different thermal characteristics, such as the collared lemur. Among the climatic variables, the CANCOR model indicates that ambient temperature was the only climatic factor strongly associated with the set of thermoregulatory behaviors (Table 2). Thus, the animals spent more time resting, huddling, and curling and less time in prone postures as temperature decreases. This result conﬁrms that ambient temperature is a good predictor of the frequencies of thermoregulatory behaviors. Conversely, humidity and wind speed did not emerge as important climatic factors, perhaps because of their relatively low ﬂuctuations in the littoral forest environment (Donati and Borgognini, 2006). Based on captive studies, the thermoneutral zone of E. fulvus, a species closely related to E. collaris, is placed between 22 and 308C (Daniels, 1984). Mean minimum temperatures recorded at Sainte Luce were below the thermoneutral zone for both the cool-wet and the transitional-dry seasons (see Fig. 1). This means that thermoregulatory tactics would be needed primarily to counteract cold stress and to maintain a high body temperature during the coolest segments of the year. Not surprisingly, given the relatively southern latitude of Sainte Luce, the CANCOR model indicates that day American Journal of Physical Anthropology 362 G. DONATI ET AL. length also had a relevant role in shaping thermoregulatory responses. This result apparently matches observations on South-African baboons, where variations in day length are one of the primary factors inﬂuencing thermoregulatory responses, since long days relax the foraging constraints over the midday periods (Hill et al., 2003; Hill, 2006). But we found an opposite trend in collared lemurs, since resting time and postural thermoregulation increase as day length decreases. This result could be explained by the strong association of day length with both temperature and phenological cycles at Sainte Luce (Bollen and Donati, 2005), making it difﬁcult to separate the effect of the former from those of the latter factors, even when using multivariate statistics. Furthermore, day length effects are difﬁcult to evaluate in a cathemeral lemur, which is not constrained by a single phase of the 24-h cycle (Curtis and Donati, in press). The CANCOR model shows that two dietary-related variables, i.e., feeding time and the availability of metabolizable energy in the diet, had very strong associations with the set of thermoregulatory behaviors. In particular, the proportion of resting and the frequency of huddling and curled postures increased as feeding time and energy from the diet decreased (Table 2). The other two dietary-related variables used in our model, fruit availability and height of feeding trees, were also correlated to the thermoregulatory behaviors, though by a weaker association if compared to energy from diet and feeding time. Fruit availability seems to behave in a similar way as feeding time; while the height of feeding trees, not surprisingly, is most associated with the time spent in the upper canopy by a positive relationship. Thus, as predicted, non climatic, dietary-related factors had a relevant role in shaping not only activity adjustments, but also social and postural thermoregulation. While diet has been often identiﬁed as a signiﬁcant factor in shaping anthropoid (Dasilva, 1992; Milton, 1998; Hill and Dunbar, 2002) and lemur activity-budget (Engqvist and Richard, 1991; Ganzhorn, 2002; Ganzhorn et al., 2003; Tarnaud, 2006; Donati et al., 2009), it has rarely been analyzed in relation to speciﬁc thermoregulatory behaviors in the ﬁeld. Dasilva (1993) has shown that postural changes of Colobus polykomos are not solely related to climate, but they also reﬂect food availability and energy content of food consumed. Pochron (2000) hypothesized that baboons could not afford micro-habitat selection during their activities in the dry season because of the constraints of low food availability. In captivity, changes of food provisioning act as a triggering mechanism for hypothermia and the occurrence of daily torpor in Microcebus spp. (Aujard et al., 1998; Genin and Perret, 2000). An important role of the diet in determining the proportion of thermoregulatory behaviors in lemurs was expected for a number of reasons. First, Malagasy lemurs are known to face long seasonal bottle-necks in terms of food availability both in the western seasonal forests (Sorg and Rohner, 1996; Bollen et al., 2005) and in the eastern rainforests (Wright, 1999; Bollen and Donati, 2005; Wright et al., 2005). These wide ﬂuctuations are likely to expose frugivorous lemurs to periodic timewindows of limited energy intake (Ganzhorn et al., 2003; but see Curtis, 2004). Some of the most striking adaptations to seasonality found in lemurs, such as torpor, have already been related to this phenomenon (Wright, 1999; Fietz and Dausmann, 2006). Second, all lemur species studied so far are hypometabolic (Genoud, 2002; Simmen et al., 2010), which could also be interpreted as American Journal of Physical Anthropology an adaptation to scarce and unpredictable resources (Kurland and Pearson, 1986; McNab, 1986). However, in captivity the low metabolic rate of Eulemur species seems to be associated with high body temperatures (Daniels, 1984; Müller, 1985), which would explain the overall high levels of activity typical of brown lemurs (Donati et al., 2007). If such a pattern holds true in the wild as well, it means that these lemurs have evolved physiological and possibly behavioral traits to keep high body temperatures in spite of their low metabolic rate. The output of our multivariate model shows that proportion of resting, huddling, and postural thermoregulation are all behaviors associated with climatic and dietaryrelated ﬂuctuations in collared lemurs. Therefore, these results suggest that lemurs rely on activity and postural adjustments, but also on social thermoregulation to compensate for seasonal environmental variations. ACKNOWLEDGMENTS This study was carried out under the collaboration agreement between the Departments of Animal Biology and Anthropology of the University of Antananarivo, the Institute of Zoology of Hamburg University and QIT Madagascar Minerals. The authors thank the Commission Tripartite of the Malagasy Government, the Ministère des Eaux et Forêts, and QMM for their collaboration and permissions to work in Madagascar. They thank Joerg Ganzhorn for his continuous scientiﬁc support. They acknowledge Manon Vincelette, JeanBaptiste Ramanamanjato, and Laurent Randriashipara for providing help at various stages of this research. They are grateful to An Bollen for providing additional data on collared lemur ecology. Many thanks to Dauphin Mbola, Givé Sambo, Ramisy Edmond, the local assistants who helped with the collection of behavioral and phenology data. Irene Tomaschewsky helped with plant analyses. LITERATURE CITED Altmann J. 1974. Observational study of behavior: sampling methods. Behavior 49:227–265. Aujard F, Perret M, Vannier G. 1998. Thermoregulatory responses to variation of photoperiod and ambient temperature in the male lesser mouse lemur: a primitive or an advanced adaptive character? J Comp Physiol B 168:540–548. Bethge P, Munks S, Otley H, Nicol S. 2004. Platypus burrow temperatures at a subalpine Tasmanian lake. Proc Linn Soc NSW 125:273–276. Bicca-Marques JC, Calegaro-Marques C. 1998. Behavioral thermoregulation in a sexually and developmentally dichromatic neotropical primate, the black-and-gold howling monkey (Alouatta caraya). Am J Phys Anthropol 106:533–546. Bollen A, Donati G. 2005. Phenology of the littoral forest of Sainte Luce, southeast Madagascar. Biotropica 37:32–43. Bollen A, Donati G, Fietz J, Schwab D, Ramanamanjato JB, Randrihasipara L, van Elsacker L, Ganzhorn JU. 2005. Fruit characteristics in a dry deciduous and a humid littoral forest of Madagascar: evidence for selection pressure through abiotic constraints rather than through coevolution by seed dispersers. In: Dew L, Boubli JP, editors. Tropical fruits and frugivores: the search for strong interactors. Dordrecht: Kluwer Academic. p 92–118. Bradley SR, Hudson JW. 1974. Temperature regulation in the tree shrew Tupaia glis. Comp Biochem Physiol A 48:55–60. Brice PH, Gordon CG, Beard LA, Donovan JA. 2002. Heat tolerance of short-beaked echidnas (Tachyglossus Aculeatus) in the ﬁeld. J Therm Biol 27:449–457. BEHAVIORAL THERMOREGULATION IN COLLARED LEMURS Campbell JL, Williams CV, Eisemann JH. 2004. Use of total dietary ﬁber across four lemur species (Propithecus verreauxi coquereli, Hapalemur griseus griseus, Varecia variegata, and Eulemur fulvus): does ﬁber type affect digestive efﬁciency? Am J Primatol 64:323–335. Campos FA, Fedigan LM. 2009. Behavioral adaptations to heat stress and water scarcity in white-faced capuchins (Cebus capucinus) in Santa Rosa National Park. Costa Rica. Am J Phys Anthropol 138:101–111. Conklin-Brittain NL, Knott CD, Wrangham RW. 2006. Energy intake by wild chimpanzees and orangutans: methodological considerations and preliminary comparison. In: Hohmann G, Robbins MM, Boesch C, editors. Feeding ecology in apes and other primates. Cambridge: Cambridge University Press. Cowlishaw G, Dunbar R. 2000. Primate conservation biology. Chicago: University of Chicago Press. Curtis DJ. 2004. Diet and nutrition in wild mongoose lemurs (Eulemur mongoz) and their implications for the evolution of female dominance and small group size in lemurs. Am J Phys Anthropol 124:234–247. Curtis DJ, Donati G. The role of photoperiod in the evolution of temporal plasticity in lemurs. In: Masters J, Gamba M, Génin F, editors. Leaping ahead: advances in prosimian biology (in press). Curtis DJ, Rasmussen MA. 2006. The evolution of cathemerality in primates and other mammals: a comparative and chronoecological approach. Folia Primatol 77:178–193. Curtis DJ, Zaramody A, Martin RD. 1999. Cathemeral activity in the mongoose lemur, Eulemur mongoz. Am J Primatol 47:279–298. Daniels HL. 1984. Oxygen consumption in Lemur fulvus: deviation from the ideal model. J Mammal 65:584–592. Dasilva GL. 1992. The western black-and-white colobus as a low-energy strategist: activity budgets, energy expenditure and energy intake. J Anim Ecol 61:79–91. Dasilva GL. 1993. Postural changes and behavioural thermoregulation in Colobus polykomos: the effect of climate and diet. Afr J Ecol 31:226–241. Dewar RE, Richard AF. 2007. Evolution in the hypervariable environment of Madagascar. Proc Natl Acad Sci USA 104:13723–13727. Dhal JF, Smith EO. 1985. Assessing variation in the social behavior of stumptailed macaques using thermal criteria. Am J Phys Anthropol 68:467–477. Donati G, Baldi N, Morelli V, Ganzhorn JU, Borgognini-Tarli SM. 2009. Proximate and ultimate determinants of cathemeral activity in brown lemurs. Anim Behav 77:317–325. Donati G, Bollen A, Borgognini-Tarli SM, Ganzhorn JU. 2007. Feeding over 24-h cycle: dietary ﬂexibility of cathemeral collared lemurs (Eulemur collaris). Behav Ecol Sociobiol 61:1237–1251. Donati G, Borgognini-Tarli SM. 2006. Inﬂuence of abiotic factors on cathemeral activity: the case of Eulemur fulvus collaris in the littoral forest of Madagascar. Folia Primatol 77:104–122. Engqvist A, Richard A. 1991. Diet as a possible determinant of cathemeral activity patterns in primates. Folia Primatol 57:169–172. Estrada A, Coates-Estrada R. 1996. Tropical rain forest fragmentation and wild populations of primates at Los Tuxtlas. Int J Primatol 5:759–783. Fietz J, Dausmann K. 2006. Big is beautiful: fat storage and hibernation as a strategy to cope with marked seasonality in the fat-tailed dwarf lemur (Cheirogaleus medius). In: Gould L, Sauther ML, editors. Lemurs: ecology and adaptation. New York: Springer. p 97–110. Ganzhorn JU. 1995. Low-level forest disturbance effects on primary production, leaf chemistry, and lemur populations. Ecology 76:2984–2096. Ganzhorn JU. 2002. Distribution of a folivorous lemur in relation to seasonally varying food resources: integrating quantitative and qualitative aspects of food characteristics. Oecologia 131:427–435. Ganzhorn JU. 2003. Habitat description and phenology. In: Setchell JM, Curtis DJ, editors. Field and laboratory methods 363 in primatology. A practical guide. Cambridge: Cambridge University Press. p 40–56. Ganzhorn JU, Andrianasolo T, Andrianjazalahatra T, Donati G, Fietz J, Lahann P, Norscia I, Rakotondranary J, Rakotondratsima BM, Ralison J, Ramarokoto REAF, Randriamanga S, Rasarimanana S, Rakotosamimanana B, Ramanamanjato JB, Randria G, Tovonanahary Rasolofoharivelo M, RazanahoeraRakotomalala M, Schmid J, Sommer S. 2007. Lemurs in evergreen littoral forest fragments. In: Ganzhorn JU, Goodman SM, Vincelette M, editors. Biodiversity, ecology, and conservation of littoral ecosystems of southeastern Madagascar, Tolagnaro (Fort Dauphin). Washington, DC: Smithsonian Institution Press. p 223–235. Ganzhorn JU, Klaus S, Ortmann S, Schmid J. 2003. Adaptation to seasonality: some primate and non-primate examples. In: Kappeler PM, Pereira ME, editors. Primate life histories and socioecology. Chicago: University of Chicago Press. p 132–148. Genin F, Perret M. 2000. Photoperiod-induced changes in energy balance in gray mouse lemurs. Physiol Behav 71:315– 321. Genoud M. 2002. Comparative studies of basal rate of metabolism in primates. Evol Anthropol 11:108–111. Grigg GC, Beard LA, Barnes JA, Perry LI, Fry GJ, Hawkins M. 2003. Body temperature in captive long-beaked echidnas (Zaglossus bartoni). Comp Biochem Phys A 136:911–916. Hasler JF, Sorenson W. 1974. Behavior of the tree shrew, Tupaia chinensis, in captivity. Am Midl Nat 91:294–314. Hanya G, Kiyono M, Hayaishi S. 2007. Behavioral thermoregulation of wild Japanese macaques: comparisons between two subpopulations. Am J Primatol 69:802–815. Hill RA. 2006. Thermal constraints on activity scheduling and habitat choice in baboons. Am J Phys Anthropol 129:242–249. Hill RA, Barret L, Gaynor D, Weingrill T, Dixon P, Payne H, Henzi SP. 2003. Day length, latitude, and behavioural (in)ﬂexibility in baboons (Papio cynocephalus ursinus). Behav Ecol Sociobiol 53:278–286. Hill RA, Dunbar RIM. 2002. Climatic determinants of diet and foraging behavior in baboons. Evol Ecol 16:579–593. Hill RA, Weingrill T, Barrett L, Henzi SP. 2004. Indices of environmental temperatures for primates in open habitats. Primates 45:7–13. Johns AD. 1986. Effects of selective logging on the behavioral ecology of West Malaysian primates. Ecology 67:684–694. Jolly A. 1966. Lemur behavior. Chicago: University of Chicago Press. Jolly A. 1984. The puzzle of female feeding priority. In: Small M, editor. Female primates: studies by women primatologists. New York: Alan R. Liss. p 197–215. Kappeler PM. 1993. Variation and social structure: the effects of sex and kinship on social interactions in three lemur species. Ethology 93:125–145. Kappeler PM. 2000. Causes and consequences of unusual sex ratios among lemurs. In: Kappeler PM, editor. Primate males: causes and consequences of variation in group composition. Cambridge: Cambridge University Press. p 55–63. Kauffman AS, Paul MJ, Butler MP, Zucker I. 2003. Huddling, locomotor, and nest-building behaviors of furred and furless Siberian hamsters. Physiol Behav 79:247–256. Körtner G, Geiser F. 2000. Torpor and activity patterns in freeranging sugar gliders Petaurus breviceps (Marsupialia). Oecologia 123:350–357. Kosheleff VP, Anderson NK. 2009. Temperature’s inﬂuence on the activity budget, terrestriality, and sun exposure of chimpanzees in the Budongo Forest, Uganda. Am J Phys Anthropol 139:172–181. Kotze J, Bennett NC, Scantlebury M. 2008. The energetics of huddling in two species of mole-rat (Rodentia: Bathyergidae). Physiol Behav 93:215–221. Kurland JA, Gaulin SJC. 1987. Comparability among measures of primate diets. Primates 28:71–77. Kurland JA, Pearson JD. 1986. Ecological signiﬁcance of hypometabolism in nonhuman primates: allometry, adaptation, and deviant diets. Am J Phys Anthropol 71:445–457. American Journal of Physical Anthropology 364 G. DONATI ET AL. Lehman SM, Rajaonson A, Day S. 2006. Edge effects and their inﬂuence on lemur density and distribution in Southeast Madagascar. Am J Phys Anthropol 129:232–241. Lovejoy TE, Bierreegaard RO Jr, Rylands AB, Malcom JR, Quintela CE, Harper LH, Brown KS Jr, Powell AH, Powell GVN, Schubart HOR, Hays MB. 1986. Edge and other effects of isolation on Amazon forest fragments. In: Soule ME, editor. Conservation biology: the science of scarcity and diversity. Sunderland, Massachusetts: Sinauer Associates. p 257–285. McCarron HCK, Buffestein R, Fanning FD, Dawson TJ. 2001. Free-ranging heart rate, body temperature and energy metabolism in eastern grey kangaroos (Macropus giganteus) and red kangaroos (Macropus rufus) in the arid regions of South East Australia. J Comp Physiol B 171:401–411. McGarigal K, Cushman S, Stafford S. 2000. Multivariate statistics for wildlife and ecology research. New York: Springer. McNab BK. 1978. The comparative energetics of neotropical marsupials. J Comp Physiol B 125:115–128. McNab BK. 1986. The inﬂuence of food habits on the energetic of eutherian mammals. Ecol Monogr 56:1–19. Milton K. 1998. Physiological ecology of howlers (Alouatta): energetic and digestive considerations and comparison with colobinae. Int J Primatol 19:513–548. Morellato LPC, Talora DC, Takahasi A, Bencke CC, Romera EC, Zipparro VB. 2000. Phenology of Atlantic rainforest trees: a comparative study. Biotropica 32:811–823. Morland HS. 1993. Seasonal behavioral variation and its relationship to thermoregulation in ruffed lemurs (Varecia variegata variegata). In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum. p 193–203. Mount LE. 1979. Adaptation to thermal environment: man and his productive animals. London: Arnold. Müller EF. 1985. Basal metabolic rates in primates: the possible role of phylogenetic and ecological factors. Comp Biochem Physiol A 81:707–711. Onderdonk DA, Chapman CA. 2000. Coping with forest fragmentation: the primates of Kibale National Park, Uganda. Int J Primatol 21:587–611. Ortmann S, Bradley BJ, Stolter C, Ganzhorn JU. 2006. Estimating the quality and composition of wild animal diets—a critical survey of methods. In: Hohmann G, Robbins MM, Boesch C, editors. Feeding ecology in apes and other primates. Ecological, physical and behavioural aspects. Cambridge: Cambridge University Press. p 395–418. Ostner J. 2002. Social thermoregulation in redfronted lemurs (Eulemur fulvus rufus). Folia Primatol 73:175–180. Overdorff DJ. 1998. Are Eulemur species pair-bonded? Social organization and mating strategies in Eulemur fulvus rufus from 1988–1995 in southeast Madagascar. Am J Phys Anthropol 105:153–166. Pereira ME, Strohecker RA, Cavigelli SA, Hughes CL, Pearson DD. 1999. Metabolic strategy and social behavior in Lemuridae. In: Rakotosamimanana B, Rasamimanana H, Ganzhorn JU, Goodman SM, editors. New directions in lemur studies. New York: Plenum. p 93–118. Pochron ST. 2000. Sun avoidance in the yellow baboons (Papio cynocephalus cynocephalus) of Ruaha National Park, Tanzania. Variations with season, behaviour and weather. Int J Biometeorol 44:141–147. Rakotondranary J. 2004. Effets de la dégradation sur la population de Avahi laniger dans la forêt littorale de Sainte-Luce. American Journal of Physical Anthropology DEA, Département de Paléontologie et d’Anthropologie Biologique. Université d’Antananarivo. Reﬁnetti R. 1998. Body temperature and behavior of tree shrews and ﬂying squirrels in a thermal gradient. Physiol Behav 63:517–520. Richard AF, Dewar RE. 1991. Lemur ecology. Annu Rev Ecol Syst 22:145–175. Rode KD, Chapman CA, McDowell LR, Stickler C. 2006. Nutritional correlates of population density across habitats and logging intensities in redtail monkeys (Cercopithecus ascanius). Biotropica 38:625–634. Rübsamen U, Hume ID, Rübsamen K. 1983. Effect of ambient temperature on autonomic thermoregulation and activity patterns in the rufous rat-kangaroo (Aepyprymnus rufescens: Marsupialia). J Comp Physiol B 153:175–179. Satinoff E. 1978. Neural organization and evolution of thermal regulation in mammals. Science 201:16–22. Scantlebury M, Bennett NC, Speakman JR, Pillay N, Schradin C. 2006. Huddling in groups lead to daily energy savings in free-living African fourstriped grass mice, Rhabdomys pumilio. Funct Ecol 20:166–173. Schino G, Troisi A. 1990. Behavioral thermoregulation in longtailed macaques: effect on social preference. Physiol Behav 47:1125–1128. Seebacher F, Franklin CE. 2005. Physiological mechanism of thermoregulation in reptiles: a review. J Comp Physiol B 175:533–541. Simmen B, Bayart F, Rasamimanana H, Zahariev A, Blanc S, Pasquet P. 2010. Total energy expenditure and body composition in two free-living sympatric lemurs. PLoS ONE 5:e9860. Sorg JP, Rohner U. 1996. Climate and tree phenology of the dry deciduous forest: the Kirindy forest. Prim Rep 46:57–80. Stelzner JK, Hausfater G. 1986. Posture, microclimate, and thermoregulation in yellow baboons. Primates 27:449–463. Stevenson RD. 1985. The relative importance of behavioural and physiological adjustments controlling body temperature in terrestrial ectotherms. Am Nat 126:362–386. Sussman RW. 1974. Ecological distinctions in sympatric species of Lemur. In: Martin RD, Doyle GA, Walker AC, editors. Prosimian biology. London: Duckworth. p 75–108. Tarnaud L. 2006. Cathemerality in the Mayotte brown lemur (Eulemur fulvus): seasonality and food quality. Folia Primatol 77:166–177. Tattersall I. 1982. The primates of Madagascar. New York: Columbia University Press. Takemoto H. 2004. Seasonal change in terrestriality of chimpanzees in relation to microclimate in the tropical forest. Am J Phys Anthropol 124:81–92. Thompson SD, Power ML, Rutledge CE, Kleiman DG. 1994. Energy metabolism and thermoregulation in the golden lion tamarin (Leontopithecus rosalia). Folia Primatol 63:131–143. van Schaik CP, Kappeler PM. 1996. The social systems of gregarious lemurs: lack of convergence with anthropoids due to evolutionary disequilibrium? Ethology 102:915–941. van Soest PJ. 1994. Nutritional ecology of the ruminants. Ithaca, NY: Cornell University Press. Wright PC. 1999. Lemur traits and Madagascar ecology: coping with an island environment. Yearb Phys Anthropol 42:31–72. Wright PC, Razaﬁndratsita VR, Pochron ST, Jernvall J. 2005. The key to Madagascar frugivores. In: Dew L, Boubli JP, editors. Tropical fruits and frugivores: the search for strong interactors. Dordrecht: Kluwer Academic. p 121–138.